INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

iHUD

-0.79

è¦ļéĨĴ

-0.72

Sleep

-0.71

 crawl

-0.68

ques

-0.64

aters

-0.63

 theaters

-0.61

 Volcano

-0.59

TeX

-0.58

athe

-0.58

POSITIVE LOGITS

imil

0.77

akuya

0.71

rontal

0.70

enture

0.68

taking

0.67

uing

0.65

piration

0.63

oyal

0.63

iked

0.63

ivities

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.